CN106845509A - A kind of Coal-rock identification method based on bent wave zone compressive features - Google Patents
A kind of Coal-rock identification method based on bent wave zone compressive features Download PDFInfo
- Publication number
- CN106845509A CN106845509A CN201610909451.6A CN201610909451A CN106845509A CN 106845509 A CN106845509 A CN 106845509A CN 201610909451 A CN201610909451 A CN 201610909451A CN 106845509 A CN106845509 A CN 106845509A
- Authority
- CN
- China
- Prior art keywords
- image
- coal
- rock
- vector
- coal petrography
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000011435 rock Substances 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 title claims abstract description 29
- 239000003245 coal Substances 0.000 claims abstract description 46
- 239000011159 matrix material Substances 0.000 claims abstract description 12
- 238000000605 extraction Methods 0.000 claims description 4
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 6
- 238000005286 illumination Methods 0.000 abstract description 4
- 238000005065 mining Methods 0.000 abstract description 3
- 238000012706 support-vector machine Methods 0.000 abstract description 3
- 230000008859 change Effects 0.000 abstract description 2
- 238000003384 imaging method Methods 0.000 abstract description 2
- 238000004364 calculation method Methods 0.000 abstract 1
- 230000004069 differentiation Effects 0.000 abstract 1
- 230000006870 function Effects 0.000 description 7
- 238000001514 detection method Methods 0.000 description 5
- 239000000428 dust Substances 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000005251 gamma ray Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of Coal-rock identification method based on bent wave zone compressive features, the method carries out multiple dimensioned warp wavelet to the coal petrography sample image for gathering, and obtains the bent wave system number vector under correspondence yardstickWithUsing calculation matrix pairWithVector carries out dimensionality reduction after concatenated in order；CascadeCoal petrography image feature vector is constituted with the vector after dimensionality reduction；With the sample image feature set Training Support Vector Machines identifier for extracting；In identification, coal petrography image feature vector to be identified is input into identifier, realizes the differentiation to coal petrography image.The image of the method coal, rock under different illumination, different points of view is used as training sample, influenceed small by illumination and imaging viewpoint change, with recognition correct rate higher and identification stability, the production processes such as cash can be selected to provide reliable coal petrography identification information for automated mining, automatic coal discharge, automation.
Description
Technical field
The present invention relates to a kind of Coal-rock identification method based on bent wave zone compressive features, belong to coal petrography identification field.
Background technology
It is coal or rock that coal petrography identification automatically identifies coal petrography object with a kind of method.In coal production process, coal
Rock identification technology can be widely applied to roller coal mining, driving, top coal caving, raw coal and select the production links such as spoil, for reducing
Getting working face operating personnel, mitigate labor strength, improve operating environment, realize safety of coal mines efficiently production have it is important
Meaning.
There are various methods to be applied to coal petrography to recognize, such as natural Gamma ray detection, radar detection, stress pick, infrared spy
Survey, active power monitoring, shock detection, sound detection, dust detection, memory cut etc., but there is problems with these methods：
1. need to install various kinds of sensors acquisition information additional on existing, cause apparatus structure complicated, high cost.2. coal-winning machine rolling
Stress is complicated in process of production, vibration is violent, serious wear, dust are big for the equipment such as cylinder, development machine, and sensor deployment is relatively stranded
Difficulty, is easily caused mechanical component, sensor and electric wiring and is damaged, and device reliability is poor.3. for different type machinery
There is larger difference, it is necessary to carry out personalized customization in equipment, the selection of optimal type and the picking up signal point of sensor, system
Bad adaptability.
To solve the above problems, image technique be also increasingly taken seriously and have developed some be based on image technique coal petrographys
Recognition methods.But though existing wavelet method can effectively describe the local letter of coal petrography with good Time-Frequency Localization ability
Breath, but wavelet transformation reflection be signal point singularity, its base is isotropic, it is impossible to accurately express coal petrography image side
The direction of edge, cannot also realize the rarefaction representation to image, therefore the important boundary curve of small echo coal petrography image beyond expression of words is special
Levy, thus also have very big deficiency on recognition correct rate.
A kind of Coal-rock identification method for solving or at least improving one or more intrinsic in the prior art problems is needed, with
Improve coal petrography discrimination and identification stability.
The content of the invention
Therefore, it is an object of the invention to provide a kind of Coal-rock identification method based on bent wave zone compressive features, the method
Regard straight line as per a bit of curve approximation, multiple dimensioned, multidirectional warp wavelet is then carried out to image again, so that phenogram picture
Curvilinear characteristic, and the recognition methods is influenceed small by illumination and imaging viewpoint change, so that the method is with identification higher
Accuracy and identification stability, can select the production processes such as cash to provide reliable coal for automated mining, automatic coal discharge, automation
Rock identification information.
It is according to a kind of embodiment form, there is provided a kind of Coal-rock identification method based on bent wave zone compressive features including as follows
Step：
A. collection size is the coal sample image set of 128 × 128 pixels respectivelyWith rock specimens image setExtract each coal image I in the coal sample image setcCharacteristics of image Xc∈RQWith the rock specimens image
Concentrate each rock image IrCharacteristics of image Xr∈RQ, composing training collectionWithEvery wherein described figure
The characteristics of image X of picturec∈RQOr Xr∈RQExtraction step is as follows：
(1) to every sample image, yardstick s=2 is used-j, j=0,1,2,3 march Wave Decomposition, obtain correspondence yardstick under
Bent wave system number vectorWith
(2) willWithConcatenated in order constitutes vector z ∈ RN×1, with random matrix ψ ∈ RM×N, M ＜ ＜ N
It is projected into lower dimensional space, i.e. P=ψ z；
(3) cascadeWith P, it isThe as characteristics of image；
B. with the sample image feature set extractedWithTraining identifierWhen sgn ()=1 is coal, sgn ()=- 1 is rock, wherein, I=1,2 ..., T,yt∈ { -1,1 },It is the constant more than 0, Ωt∈RQ, setObtained in training with σ；
C. for coal petrography image to be identified, characteristics of image X is extracted with step A identicals method, is input to and uses step B
The identifier of foundation is differentiated.
Wherein, it is 0 that the element in the random matrix ψ in step A extraction process (2) obeys average, and variance is the Gauss of 1/M
Distribution.
Brief description of the drawings
By following explanation, accompanying drawing embodiment becomes apparent, and it is only preferred with least one being described in conjunction with the accompanying
But the way of example of non-limiting example is given.
Fig. 1 is the flow chart of Coal-rock identification method of the present invention.
Specific embodiment
The invention discloses a kind of Coal-rock identification method based on bent wave zone compressive features, Fig. 1 is based on bent wave zone compression
The flow chart of the Coal-rock identification method of feature, reference picture 1, the specific implementation steps of one embodiment of the present of invention are as follows：
If A. gathering the dry coal and rock of different illumination, different points of view from the scene of coal petrography identification mission such as coal-face
Sample image, therefrom interception only comprising coal and only comprising rock pixel size for 128 × 128 image-region, if color
Color image, then be first converted into gray level image with Formulas I=0.299R+0.587G+0.114B；Coal, rock specimens after treatment
Image respectively constitutes coal sample image setWith rock specimens image setThe coal sample image set is extracted respectively
In each coal image IcCharacteristics of image Xc∈RQWith each rock image I in the rock specimens image setrCharacteristics of image
Xr∈RQ, composing training collectionWith
Every characteristics of image Xx ∈ R of imageQOr Xr∈RQExtraction step is as follows：
(1) to every sample image, yardstick s=2 is used-j, j=0,1,2,3 march Wave Decomposition, obtain correspondence yardstick under
Bent wave system number LJ=0、HJ=1、HJ=2And HJ=3, the bent wave system number under correspondence yardstick is cascaded by row successively respectively, obtain bent wave system number
VectorWith
Wherein, LJ=0It is 21 × 21 coefficient matrix, HJ=11 be made up of the matrix of 19 × 43 or 43 × 19 sizes ×
8cell coefficient matrixes, HJ=21 × 16cell the systems being made up of the matrix of 35 × 44 or 44 × 35 or 32 × 42 or 42 × 32 sizes
Matrix number, HJ=3It is 128 × 128 coefficient matrix.
(2) willWithConcatenated in order constitutes vector z ∈ RN×1, with random matrix ψ ∈ RM×N, M ＜ ＜ N
It is projected into lower dimensional space, i.e. P=ψ z；
Wherein, random matrix ψ needs to meet constraint isometry (Restricted Isometry Property, RIP), ψ
In element obey average be 0, variance for 1/M Gaussian Profile；
In the present invention, M=20 is taken.
(3) cascadeWith P, it isThe as characteristics of image.
B. selection SVMs is that the input of the feature of multidimensional can be mapped into height carrying out the advantage of tagsort
The nuclear space of dimension, so that inseparable data obtain new feature originally, advantageously in classification.
With the sample image feature set extractedWithThe classification Nonlinear Support Vector Machines identification of training two
Device, specifically includes following steps：
(1) it is W to set optimizing decision face equationTXi+ b=0, wherein W are weight vector, and b is bias, XiIt is i-th sample
Characteristics of image, the object function of solution is：
Constraints is yi(WTXi+b)≥1-ξi, ξi>=0, i=1,2,3 ...,
K, wherein C are penalty coefficient, ξiIt is the slack variable under the conditions of linearly inseparable, yi∈ { -1,1 } is category label.
(2) Lagrange multiplier methods are utilized, object function can be converted into following constrained optimization problem：
WhereinIt is Lagrange multipliers, meets about
Beam condition：0≤αi≤ C,
(3) optimal solution is tried to achieveOptimizing decision faceKnow
Other device isWhereinI=1,2 ..., T, support to
Quantity setThe set of corresponding feature Ω t after the training sample rearrangement of i=1,2,3 ..., K；
It is kernel function, meets Mercer theorems, common kernel function has two kinds of multinomial and radial direction base；
In the present invention,Using Radial basis kernel function (RBF),σ is
Kernel function width.
In order to be supported the selection of vector machine parameter, optimal penalty coefficient C and kernel function width cs are selected, sample will be trained
Originally 5-fold cross validations are carried out, based on grid data service (Grid Search), the hunting zone of C is [2-2, 230], σ's searches
Rope scope is [2-26, 20], step-length is 1, searches optimal ginseng of the coal petrography classification accuracy rate highest parameter combination as model
Number.According to parameter selection result, determine that optimized parameter C is 0.0625 for 2, σ.
C. for coal petrography image I to be identifiedX, pixel size of the interception only comprising coal or rock object is 128 × 128 figure
As region, characteristics of image X is extracted with step A identicals method, be input to the identifier set up with step BDifferentiated, when sgn ()=1 is coal, sgn ()=- 1 is rock.
Claims (2)
1. a kind of Coal-rock identification method based on bent wave zone compressive features, it is characterised in that comprise the following steps：
A. collection size is the coal sample image set of 128 × 128 pixels respectivelyWith rock specimens image setCarry
Take each coal image I in the coal sample image setcCharacteristics of image Xc∈RQWith each in the rock specimens image set
Rock image IrCharacteristics of image Xr∈RQ, composing training collectionWithThe image of every wherein described image is special
Levy Xc∈RQOr Xr∈RQExtraction step is as follows：
(1) to every sample image, yardstick s=2 is used-j, j=0,1,2,3 march Wave Decomposition, obtain correspondence yardstick under Qu Bo
Coefficient vectorWith
(2) willWithConcatenated in order constitutes vector z ∈ RN×1, with random matrix ψ ∈ RM×N, M ＜ ＜ N are thrown
Shadow is to lower dimensional space, i.e. P=ψ z；
(3) cascadeWith P, it isThe as characteristics of image；
B. with the sample image feature set extractedWithTraining identifierWhen sgn ()=1 is coal, sgn ()=- 1 is rock, wherein, For
Constant more than 0, Ωt∈RQ, setObtained in training with parameter σ；
C. for coal petrography image to be identified, characteristics of image X is extracted with step A identicals method, is input to and is set up with step B
Identifier differentiated.
2. the Coal-rock identification method based on bent wave zone compressive features according to claim 1, it is characterised in that in step A with
It is 0 that element in machine matrix ψ obeys average, and variance is the Gaussian Profile of 1/M.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610909451.6A CN106845509A (en) | 2016-10-19 | 2016-10-19 | A kind of Coal-rock identification method based on bent wave zone compressive features |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610909451.6A CN106845509A (en) | 2016-10-19 | 2016-10-19 | A kind of Coal-rock identification method based on bent wave zone compressive features |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106845509A true CN106845509A (en) | 2017-06-13 |
Family
ID=59146004
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610909451.6A Pending CN106845509A (en) | 2016-10-19 | 2016-10-19 | A kind of Coal-rock identification method based on bent wave zone compressive features |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106845509A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020062470A1 (en) * | 2018-09-28 | 2020-04-02 | 中国矿业大学 | Apparatus and method for recognizing coal-rock interface based on solid-state laser radar imaging |
| CN111259095A (en) * | 2020-01-08 | 2020-06-09 | 京工博创（北京）科技有限公司 | Method, device and equipment for calculating boundary of ore rock |
| CN111779524A (en) * | 2020-06-30 | 2020-10-16 | 中国矿业大学 | Intelligent coal caving method for hydraulic support of fully mechanized caving face |
| CN112200813A (en) * | 2020-09-30 | 2021-01-08 | 中国矿业大学(北京) | Coal and gangue identification method and system considering illumination factor |
| CN113052208A (en) * | 2021-03-10 | 2021-06-29 | 神华神东煤炭集团有限责任公司 | Coal rock identification method based on vision, storage medium and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120294543A1 (en) * | 2009-12-04 | 2012-11-22 | Pradeep Sen | System and methods of compressed sensing as applied to computer graphics and computer imaging |
| CN102902950A (en) * | 2012-06-26 | 2013-01-30 | 西安电子科技大学 | Human movement identification method based on second-generation Curvelet transformation |
| CN103942576A (en) * | 2014-04-09 | 2014-07-23 | 中国矿业大学(北京) | Method for identifying coal and rock through airspace multiscale random characteristics |
-
2016
- 2016-10-19 CN CN201610909451.6A patent/CN106845509A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120294543A1 (en) * | 2009-12-04 | 2012-11-22 | Pradeep Sen | System and methods of compressed sensing as applied to computer graphics and computer imaging |
| CN102902950A (en) * | 2012-06-26 | 2013-01-30 | 西安电子科技大学 | Human movement identification method based on second-generation Curvelet transformation |
| CN103942576A (en) * | 2014-04-09 | 2014-07-23 | 中国矿业大学(北京) | Method for identifying coal and rock through airspace multiscale random characteristics |
Non-Patent Citations (4)
| Title |
|---|
| 刘康明: "曲波变换和最小二乘支持向量机的图像压缩算法", 《激光杂志》 * |
| 张九龙等: "基于Curvelet的人脸识别", 《计算机工程与应用》 * |
| 王新春等: "基于曲波变换决策融合的掌纹识别", 《楚雄师范学院学报》 * |
| 许学斌等: "基于离散曲波变换和支持向量机的掌纹识别方法", 《红外与毫米波学报》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020062470A1 (en) * | 2018-09-28 | 2020-04-02 | 中国矿业大学 | Apparatus and method for recognizing coal-rock interface based on solid-state laser radar imaging |
| CN111259095A (en) * | 2020-01-08 | 2020-06-09 | 京工博创（北京）科技有限公司 | Method, device and equipment for calculating boundary of ore rock |
| CN111779524A (en) * | 2020-06-30 | 2020-10-16 | 中国矿业大学 | Intelligent coal caving method for hydraulic support of fully mechanized caving face |
| CN112200813A (en) * | 2020-09-30 | 2021-01-08 | 中国矿业大学(北京) | Coal and gangue identification method and system considering illumination factor |
| CN112200813B (en) * | 2020-09-30 | 2024-02-06 | 中国矿业大学(北京) | Coal gangue identification method and system considering illumination factors |
| CN113052208A (en) * | 2021-03-10 | 2021-06-29 | 神华神东煤炭集团有限责任公司 | Coal rock identification method based on vision, storage medium and electronic equipment |
| CN113052208B (en) * | 2021-03-10 | 2023-08-25 | 神华神东煤炭集团有限责任公司 | Vision-based coal rock identification method, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106845509A (en) | A kind of Coal-rock identification method based on bent wave zone compressive features | |
| CN110390691B (en) | Ore dimension measuring method based on deep learning and application system | |
| CN104751192B (en) | Coal-rock identification method based on image block symbiosis feature | |
| CN103927553B (en) | Coal and rock recognition method based on multi-scale micro-lamination and contrast ratio joint distribution | |
| CN111161292B (en) | Ore scale measurement method and application system | |
| CN112001878A (en) | Deep learning ore scale measuring method based on binarization neural network and application system | |
| CN108197630A (en) | A kind of Coal-rock identification method based on self study | |
| CN115797798A (en) | Ecological restoration effect evaluation method based on abandoned mine remote sensing image | |
| CN111626092A (en) | Unmanned aerial vehicle image ground crack identification and extraction method based on machine learning | |
| CN107862675A (en) | A kind of real-time vision detection method for electric bucket tooth missing | |
| CN110675343B (en) | Image enhancement method for underground coal conveying image | |
| CN103500449A (en) | Satellite visible light remote sensing image cloud detection method | |
| CN103927528A (en) | Coal and rock recognition method based on close neighborhood pixel gray level joint distribution characteristics | |
| Fortin et al. | Instance segmentation for autonomous log grasping in forestry operations | |
| CN117437600A (en) | Coal flow monitoring system based on image recognition technology | |
| CN103942576B (en) | A kind of method that multiple dimensioned random character in use spatial domain recognizes coal petrography | |
| CN105243401A (en) | Coal rock recognition method based on coal structure element study | |
| CN113128388B (en) | Optical remote sensing image change detection method based on space-time spectrum characteristics | |
| CN105350963B (en) | A kind of Coal-rock identification method learnt based on relativity measurement | |
| CN104463098B (en) | With the structure tensor direction histogram feature recognition coal petrography of image | |
| CN104778461A (en) | Coal-rock recognition method based on similar measure learning | |
| CN108154076A (en) | A kind of cascade rail pedestrian detection method of machine learning algorithm | |
| CN109829511B (en) | Texture classification-based method for detecting cloud layer area in downward-looking infrared image | |
| CN110751150A (en) | FPGA-based binary neural network license plate recognition method and system | |
| CN116311205A (en) | License plate recognition method, license plate recognition device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170613 |
|
| WD01 | Invention patent application deemed withdrawn after publication |